Spaces:
Runtime error
Runtime error
Moustafa1111111111
commited on
Commit
·
7651129
1
Parent(s):
0086768
Add all application files and TTS submodule
Browse files- .gitattributes +1 -0
- .gitignore +8 -0
- Dockerfile +85 -0
- README.md +5 -4
- index.html +19 -0
- local_server_new.py +136 -0
- requirements.txt +178 -0
- runtime.txt +1 -0
- script.js +38 -0
- speaker_reference.wav +3 -0
- start.sh +13 -0
- style.css +76 -0
- web/index.html +19 -0
- web/script.js +38 -0
- web/style.css +76 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
speaker_reference.wav filter=lfs diff=lfs merge=lfs -text
|
.gitignore
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
TTS/**/*.wav
|
2 |
+
TTS/**/*.png
|
3 |
+
TTS/**/*.gif
|
4 |
+
TTS/**/*.exe
|
5 |
+
TTS/new_venv1/
|
6 |
+
__pycache__/
|
7 |
+
*.pyc
|
8 |
+
*.log
|
Dockerfile
ADDED
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
FROM python:3.9-slim-buster
|
2 |
+
|
3 |
+
# Install Git (if not already included)
|
4 |
+
RUN apt-get update && apt-get install -y git
|
5 |
+
|
6 |
+
# Set the working directory in the container
|
7 |
+
WORKDIR /app
|
8 |
+
|
9 |
+
# Copy your entire repository content (including .git, .gitmodules, TTS as a submodule pointer)
|
10 |
+
COPY . /app/
|
11 |
+
|
12 |
+
# Install build tools and other dependencies BEFORE submodule init
|
13 |
+
RUN apt-get update && apt-get install -y --no-install-recommends \
|
14 |
+
build-essential \
|
15 |
+
cmake \
|
16 |
+
pkg-config \
|
17 |
+
libblis-dev \
|
18 |
+
python3-venv \
|
19 |
+
python3-dev \
|
20 |
+
wget \
|
21 |
+
libopenblas-dev
|
22 |
+
|
23 |
+
# Initialize and update submodules
|
24 |
+
RUN git submodule init
|
25 |
+
RUN git submodule update
|
26 |
+
|
27 |
+
# Set the working directory to the TTS directory AFTER updating submodules
|
28 |
+
WORKDIR /app/TTS
|
29 |
+
|
30 |
+
# Set a generic architecture flag
|
31 |
+
ENV BLIS_ARCH="generic"
|
32 |
+
|
33 |
+
# Try to agree to the Coqui TTS license via environment variable
|
34 |
+
ENV COQUI_TTS_AGREED=1
|
35 |
+
|
36 |
+
# Create a virtual environment
|
37 |
+
RUN python3 -m venv venv
|
38 |
+
RUN . /app/venv/bin/activate
|
39 |
+
|
40 |
+
# Install Coqui TTS requirements
|
41 |
+
RUN pip install -r requirements.txt --timeout=300
|
42 |
+
|
43 |
+
# Explicitly install the TTS package itself in editable mode
|
44 |
+
RUN pip install -e . --timeout=300
|
45 |
+
|
46 |
+
# Change working directory back to /app
|
47 |
+
WORKDIR /app
|
48 |
+
|
49 |
+
# Create the model directory
|
50 |
+
RUN mkdir -p /app/models/xtts_v2
|
51 |
+
|
52 |
+
# Download XTTS v2 model files
|
53 |
+
RUN wget -O /app/models/xtts_v2/config.json https://huggingface.co/coqui/XTTS-v2/resolve/main/config.json?download=true
|
54 |
+
RUN wget -O /app/models/xtts_v2/model.pth https://huggingface.co/coqui/XTTS-v2/resolve/main/model.pth?download=true
|
55 |
+
RUN wget -O /app/models/xtts_v2/vocab.json https://huggingface.co/coqui/XTTS-v2/resolve/main/vocab.json?download=true
|
56 |
+
RUN wget -O /app/models/xtts_v2/dvae.pth https://huggingface.co/coqui/XTTS-v2/resolve/main/dvae.pth?download=true
|
57 |
+
RUN wget -O /app/models/xtts_v2/speakers_xtts.pth https://huggingface.co/coqui/XTTS-v2/resolve/main/speakers_xtts.pth?download=true
|
58 |
+
|
59 |
+
# Create the audio directory if it doesn't exist
|
60 |
+
RUN mkdir -p /app/audio
|
61 |
+
|
62 |
+
# Copy the speaker_reference.wav file if it exists at the root
|
63 |
+
COPY speaker_reference.wav /app/audio/speaker_reference.wav
|
64 |
+
|
65 |
+
# Copy the web page files
|
66 |
+
COPY web /app/web
|
67 |
+
|
68 |
+
# Copy the application code
|
69 |
+
COPY local_server_new.py /app/
|
70 |
+
|
71 |
+
# Install your other dependencies (fastapi, uvicorn, bangla, etc.)
|
72 |
+
COPY requirements.txt /app/
|
73 |
+
RUN . /app/venv/bin/activate && pip install -r /app/requirements.txt --no-cache-dir --timeout=300
|
74 |
+
|
75 |
+
# Create start.sh script
|
76 |
+
RUN echo "#!/bin/bash" > start.sh && \
|
77 |
+
echo "source /app/venv/bin/activate" >> start.sh && \
|
78 |
+
echo "/app/venv/bin/python -m uvicorn local_server_new:app --host 0.0.0.0 --port 80" >> start.sh && \
|
79 |
+
chmod +x start.sh
|
80 |
+
|
81 |
+
# Expose port
|
82 |
+
EXPOSE 80
|
83 |
+
|
84 |
+
# Run the app using the script
|
85 |
+
CMD ["./start.sh"]
|
README.md
CHANGED
@@ -1,11 +1,12 @@
|
|
1 |
---
|
2 |
-
title:
|
3 |
-
emoji:
|
4 |
-
colorFrom:
|
5 |
-
colorTo:
|
6 |
sdk: docker
|
7 |
pinned: false
|
8 |
license: apache-2.0
|
|
|
9 |
---
|
10 |
|
11 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
1 |
---
|
2 |
+
title: TTS
|
3 |
+
emoji: 😻
|
4 |
+
colorFrom: pink
|
5 |
+
colorTo: purple
|
6 |
sdk: docker
|
7 |
pinned: false
|
8 |
license: apache-2.0
|
9 |
+
short_description: XTTS Large Language Model
|
10 |
---
|
11 |
|
12 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
index.html
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<!DOCTYPE html>
|
2 |
+
<html>
|
3 |
+
<head>
|
4 |
+
<title>Text to Speech</title>
|
5 |
+
<link rel="stylesheet" href="style.css">
|
6 |
+
</head>
|
7 |
+
<body>
|
8 |
+
<h1>Text to Speech</h1>
|
9 |
+
<textarea id="inputText" rows="5" cols="50"></textarea><br><br>
|
10 |
+
<button id="convertButton">Convert to Speech</button>
|
11 |
+
<div id="status"></div>
|
12 |
+
<div id="audioOutput" style="margin-top: 20px;">
|
13 |
+
<a id="downloadLink" href="#" download="output.wav" style="display: none;">Download Audio</a>
|
14 |
+
<audio id="audioPlayer" controls style="display: none;"></audio>
|
15 |
+
</div>
|
16 |
+
|
17 |
+
<script src="script.js"></script>
|
18 |
+
</body>
|
19 |
+
</html>
|
local_server_new.py
ADDED
@@ -0,0 +1,136 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from fastapi import FastAPI
|
2 |
+
from fastapi.middleware.cors import CORSMiddleware
|
3 |
+
from pydantic import BaseModel
|
4 |
+
from fastapi.responses import FileResponse
|
5 |
+
from fastapi.staticfiles import StaticFiles
|
6 |
+
import logging
|
7 |
+
import torch
|
8 |
+
import os
|
9 |
+
from TTS.api import TTS
|
10 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
|
11 |
+
from langdetect import detect
|
12 |
+
|
13 |
+
# Allowlist XttsConfig so torch.load doesn't raise UnpicklingError
|
14 |
+
from torch.serialization import add_safe_globals
|
15 |
+
from TTS.tts.configs.xtts_config import XttsConfig
|
16 |
+
add_safe_globals([XttsConfig])
|
17 |
+
|
18 |
+
# ✅ Monkey-patch torch.load to always use weights_only=False
|
19 |
+
_original_torch_load = torch.load
|
20 |
+
def patched_torch_load(*args, **kwargs):
|
21 |
+
kwargs["weights_only"] = False
|
22 |
+
return _original_torch_load(*args, **kwargs)
|
23 |
+
torch.load = patched_torch_load
|
24 |
+
|
25 |
+
logging.basicConfig(level=logging.DEBUG)
|
26 |
+
|
27 |
+
# Initialize FastAPI
|
28 |
+
app = FastAPI()
|
29 |
+
app.add_middleware(
|
30 |
+
CORSMiddleware,
|
31 |
+
allow_origins=["*"],
|
32 |
+
allow_methods=["*"],
|
33 |
+
allow_headers=["*"],
|
34 |
+
)
|
35 |
+
|
36 |
+
# Load TTS model from local files
|
37 |
+
try:
|
38 |
+
model_dir = "/app/models/xtts_v2"
|
39 |
+
config_path = os.path.join(model_dir, "config.json")
|
40 |
+
# When providing config_path, TTS might expect the directory for model_path
|
41 |
+
tts = TTS(model_path=model_dir, config_path=config_path).to("cuda" if torch.cuda.is_available() else "cpu")
|
42 |
+
print("XTTS v2 model loaded successfully from local files.")
|
43 |
+
except Exception as e:
|
44 |
+
print(f"Error loading XTTS v2 model from local files: {e}")
|
45 |
+
print("Falling back to loading by model name (license might be required).")
|
46 |
+
tts = TTS("tts_models/multilingual/multi-dataset-xtts_v2").to("cuda" if torch.cuda.is_available() else "cpu")
|
47 |
+
|
48 |
+
# Load sentiment models
|
49 |
+
arabic_model_name = "aubmindlab/bert-base-arabertv02-twitter"
|
50 |
+
sentiment_tokenizer = AutoTokenizer.from_pretrained(arabic_model_name)
|
51 |
+
sentiment_model = AutoModelForSequenceClassification.from_pretrained("UBC-NLP/MARBERT")
|
52 |
+
sentiment_analyzer = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english")
|
53 |
+
|
54 |
+
# Input class for POST body
|
55 |
+
class Message(BaseModel):
|
56 |
+
text: str
|
57 |
+
|
58 |
+
# Language detection
|
59 |
+
def detect_language_safely(text):
|
60 |
+
try:
|
61 |
+
if any('\u0600' <= c <= '\u06FF' for c in text):
|
62 |
+
return "ar"
|
63 |
+
return detect(text)
|
64 |
+
except:
|
65 |
+
return "ar" if any('\u0600' <= c <= '\u06FF' for c in text) else "en"
|
66 |
+
|
67 |
+
# Sentiment to emotion mapping
|
68 |
+
def map_sentiment_to_emotion(sentiment, language="en"):
|
69 |
+
if language == "ar":
|
70 |
+
return "happy" if sentiment == "positive" else "sad" if sentiment == "negative" else "neutral"
|
71 |
+
return "happy" if "positive" in sentiment.lower() else "sad" if "negative" in sentiment.lower() else "neutral"
|
72 |
+
|
73 |
+
# Simple Arabic sentiment analysis
|
74 |
+
def arabic_sentiment_analysis(text):
|
75 |
+
pos_words = ["سعيد", "فرح", "ممتاز", "رائع", "جيد", "حب", "جميل", "نجاح", "أحسنت", "شكرا"]
|
76 |
+
neg_words = ["حزين", "غاضب", "سيء", "فشل", "خطأ", "مشكلة", "صعب", "لا أحب", "سخيف", "مؤسف"]
|
77 |
+
pos_count = sum(1 for word in pos_words if word in text.lower())
|
78 |
+
neg_count = sum(1 for word in neg_words if word in text.lower())
|
79 |
+
|
80 |
+
if pos_count > neg_count:
|
81 |
+
return "positive"
|
82 |
+
elif neg_count > pos_count:
|
83 |
+
return "negative"
|
84 |
+
else:
|
85 |
+
try:
|
86 |
+
inputs = sentiment_tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=128)
|
87 |
+
outputs = sentiment_model(**inputs)
|
88 |
+
sentiment_class = torch.argmax(outputs.logits).item()
|
89 |
+
return ["negative", "neutral", "positive"][sentiment_class]
|
90 |
+
except:
|
91 |
+
return "neutral"
|
92 |
+
|
93 |
+
# Main TTS endpoint
|
94 |
+
@app.post("/text-to-speech/")
|
95 |
+
def text_to_speech(msg: Message):
|
96 |
+
text = msg.text
|
97 |
+
language = detect_language_safely(text)
|
98 |
+
emotion = "neutral"
|
99 |
+
|
100 |
+
if language == "en":
|
101 |
+
try:
|
102 |
+
sentiment_result = sentiment_analyzer(text)[0]
|
103 |
+
emotion = map_sentiment_to_emotion(sentiment_result["label"])
|
104 |
+
except:
|
105 |
+
pass
|
106 |
+
else:
|
107 |
+
try:
|
108 |
+
sentiment_result = arabic_sentiment_analysis(text)
|
109 |
+
emotion = map_sentiment_to_emotion(sentiment_result, language="ar")
|
110 |
+
except:
|
111 |
+
pass
|
112 |
+
|
113 |
+
output_filename = "output.wav"
|
114 |
+
try:
|
115 |
+
tts.tts_to_file(
|
116 |
+
text=text,
|
117 |
+
file_path=output_filename,
|
118 |
+
emotion=emotion,
|
119 |
+
speaker_wav="/app/audio/speaker_reference.wav", # Updated path
|
120 |
+
language=language
|
121 |
+
)
|
122 |
+
return {
|
123 |
+
"status": "success",
|
124 |
+
"audio_file": output_filename,
|
125 |
+
"url": "/audio"
|
126 |
+
}
|
127 |
+
except Exception as e:
|
128 |
+
return {"status": "error", "message": str(e)}
|
129 |
+
|
130 |
+
# ✅ Serve the audio file
|
131 |
+
@app.get("/audio")
|
132 |
+
def get_audio():
|
133 |
+
return FileResponse("output.wav", media_type="audio/wav", filename="output.wav")
|
134 |
+
|
135 |
+
# Serve static files (your web page) from the 'web' directory
|
136 |
+
app.mount("/", StaticFiles(directory="web", html=True), name="static")
|
requirements.txt
ADDED
@@ -0,0 +1,178 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# core deps
|
2 |
+
numpy==1.22.0;python_version<="3.10"
|
3 |
+
numpy>=1.24.3;python_version>"3.10"
|
4 |
+
cython>=0.29.30
|
5 |
+
scipy>=1.11.2
|
6 |
+
langdetect
|
7 |
+
torch>=2.1
|
8 |
+
torchaudio==2.6.0
|
9 |
+
soundfile>=0.12.0
|
10 |
+
librosa>=0.10.0
|
11 |
+
scikit-learn>=1.3.0
|
12 |
+
numba==0.55.1;python_version<"3.9"
|
13 |
+
numba>=0.57.0;python_version>="3.9"
|
14 |
+
inflect>=7.5.0
|
15 |
+
tqdm>=4.67.1
|
16 |
+
anyascii>=0.3.2
|
17 |
+
pyyaml>=6.0.2
|
18 |
+
fsspec>=2025.3.2
|
19 |
+
aiohttp>=3.8.1
|
20 |
+
packaging>=24.2
|
21 |
+
mutagen==1.47.0
|
22 |
+
# deps for examples
|
23 |
+
flask>=3.1.0
|
24 |
+
# deps for inference
|
25 |
+
pysbd>=0.3.4
|
26 |
+
# deps for notebooks
|
27 |
+
umap-learn>=0.5.7
|
28 |
+
pandas>=1.4,<2.0
|
29 |
+
# deps for training
|
30 |
+
matplotlib>=3.8.4
|
31 |
+
# coqui stack
|
32 |
+
trainer>=0.0.36
|
33 |
+
# config management
|
34 |
+
coqpit>=0.0.17
|
35 |
+
# chinese g2p deps
|
36 |
+
jieba==0.42.1
|
37 |
+
pypinyin==0.54.0
|
38 |
+
# korean
|
39 |
+
hangul-romanize==0.1.0
|
40 |
+
# gruut+supported langs
|
41 |
+
gruut[de,es,fr]==2.2.3
|
42 |
+
gruut-ipa==0.13.0
|
43 |
+
gruut_lang_de==2.0.1
|
44 |
+
gruut_lang_en==2.0.1
|
45 |
+
gruut_lang_es==2.0.1
|
46 |
+
gruut_lang_fr==2.0.2
|
47 |
+
# deps for korean
|
48 |
+
jamo==0.4.1
|
49 |
+
nltk==3.9.1
|
50 |
+
g2pkk>=0.1.2
|
51 |
+
# deps for bangla
|
52 |
+
bangla==0.0.1
|
53 |
+
bnnumerizer==0.0.2
|
54 |
+
bnunicodenormalizer==0.1.7
|
55 |
+
#deps for tortoise
|
56 |
+
einops==0.8.1
|
57 |
+
transformers==4.51.2
|
58 |
+
#deps for bark
|
59 |
+
encodec==0.1.1
|
60 |
+
# deps for XTTS
|
61 |
+
unidecode>=1.3.8
|
62 |
+
num2words==0.5.14
|
63 |
+
spacy[ja]>=3
|
64 |
+
# Additional dependencies from TTS requirements
|
65 |
+
absl-py==2.2.2
|
66 |
+
aiohappyeyeballs==2.6.1
|
67 |
+
aiosignal==1.3.2
|
68 |
+
annotated-types==0.7.0
|
69 |
+
anyio==4.9.0
|
70 |
+
async-timeout==5.0.1
|
71 |
+
attrs==25.3.0
|
72 |
+
audioread==3.0.1
|
73 |
+
babel==2.17.0
|
74 |
+
blinker==1.9.0
|
75 |
+
catalogue==2.0.10
|
76 |
+
certifi==2025.1.31
|
77 |
+
cffi==1.17.1
|
78 |
+
charset-normalizer==3.4.1
|
79 |
+
click==8.1.8
|
80 |
+
cloudpathlib==0.21.0
|
81 |
+
colorama==0.4.6
|
82 |
+
comtypes==1.4.10
|
83 |
+
confection==0.1.5
|
84 |
+
contourpy==1.2.1
|
85 |
+
cycler==0.12.1
|
86 |
+
cymem==2.0.11
|
87 |
+
Cython==3.0.12
|
88 |
+
dateparser==1.1.8
|
89 |
+
decorator==5.2.1
|
90 |
+
docopt==0.6.2
|
91 |
+
fastapi==0.109.2
|
92 |
+
filelock==3.18.0
|
93 |
+
fonttools==4.57.0
|
94 |
+
frozenlist==1.5.0
|
95 |
+
grpcio==1.71.0
|
96 |
+
h11==0.14.0
|
97 |
+
huggingface-hub==0.30.2
|
98 |
+
idna==3.10
|
99 |
+
importlib_metadata==8.6.1
|
100 |
+
importlib_resources==6.5.2
|
101 |
+
itsdangerous==2.2.0
|
102 |
+
Jinja2==3.1.6
|
103 |
+
joblib==1.4.2
|
104 |
+
jsonlines==1.2.0
|
105 |
+
kiwisolver==1.4.7
|
106 |
+
langcodes==3.5.0
|
107 |
+
language_data==1.3.0
|
108 |
+
lazy_loader==0.4
|
109 |
+
llvmlite==0.43.0
|
110 |
+
marisa-trie==1.2.1
|
111 |
+
Markdown==3.8
|
112 |
+
markdown-it-py==3.0.0
|
113 |
+
MarkupSafe==3.0.2
|
114 |
+
mdurl==0.1.2
|
115 |
+
more-itertools==10.6.0
|
116 |
+
mpmath==1.3.0
|
117 |
+
msgpack==1.1.0
|
118 |
+
multidict==6.4.3
|
119 |
+
murmurhash==1.0.12
|
120 |
+
networkx==2.8.8
|
121 |
+
num2words==0.5.14
|
122 |
+
packaging
|
123 |
+
pandas>=1.4,<2.0
|
124 |
+
pillow==11.1.0
|
125 |
+
platformdirs==4.3.7
|
126 |
+
pooch==1.8.2
|
127 |
+
preshed==3.0.9
|
128 |
+
propcache==0.3.1
|
129 |
+
protobuf==6.30.2
|
130 |
+
psutil==7.0.0
|
131 |
+
pycparser==2.22
|
132 |
+
pydantic==1.10.21
|
133 |
+
pydantic_core==2.33.1
|
134 |
+
Pygments==2.19.1
|
135 |
+
pynndescent==0.5.13
|
136 |
+
pyparsing==3.2.3
|
137 |
+
python-crfsuite==0.9.11
|
138 |
+
python-dateutil==2.9.0.post0
|
139 |
+
pyttsx3==2.98
|
140 |
+
pytz==2025.2
|
141 |
+
regex==2024.11.6
|
142 |
+
requests==2.32.3
|
143 |
+
rich==14.0.0
|
144 |
+
safetensors==0.5.3
|
145 |
+
shellingham==1.5.4
|
146 |
+
six==1.17.0
|
147 |
+
smart-open==7.1.0
|
148 |
+
sniffio==1.3.1
|
149 |
+
soxr==0.5.0.post1
|
150 |
+
spacy-legacy==3.0.12
|
151 |
+
spacy-loggers==1.0.5
|
152 |
+
SpeechRecognition==3.14.2
|
153 |
+
srsly==2.5.1
|
154 |
+
starlette==0.36.3
|
155 |
+
SudachiDict-core==20250129
|
156 |
+
SudachiPy==0.6.10
|
157 |
+
sympy==1.13.1
|
158 |
+
tensorboard==2.19.0
|
159 |
+
tensorboard-data-server==0.7.2
|
160 |
+
thinc==8.3.4
|
161 |
+
threadpoolctl==3.6.0
|
162 |
+
tokenizers==0.21.1
|
163 |
+
typeguard==4.4.2
|
164 |
+
typer==0.15.2
|
165 |
+
typing-inspection==0.4.0
|
166 |
+
typing_extensions==4.13.2
|
167 |
+
tzdata==2025.2
|
168 |
+
tzlocal==5.3.1
|
169 |
+
Unidecode==1.3.8
|
170 |
+
urllib3==2.4.0
|
171 |
+
uvicorn==0.34.0
|
172 |
+
wasabi==1.1.3
|
173 |
+
weasel==0.4.1
|
174 |
+
Werkzeug==3.1.3
|
175 |
+
wrapt==1.17.2
|
176 |
+
yarl==1.19.0
|
177 |
+
zipp==3.21.0
|
178 |
+
# Force rebuild
|
runtime.txt
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
python-3.9
|
script.js
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
document.addEventListener('DOMContentLoaded', () => {
|
2 |
+
const convertButton = document.getElementById('convertButton');
|
3 |
+
const inputText = document.getElementById('inputText');
|
4 |
+
const statusDiv = document.getElementById('status');
|
5 |
+
const downloadLink = document.getElementById('downloadLink');
|
6 |
+
const audioPlayer = document.getElementById('audioPlayer');
|
7 |
+
|
8 |
+
convertButton.addEventListener('click', async () => {
|
9 |
+
const text = inputText.value;
|
10 |
+
statusDiv.textContent = 'Processing...';
|
11 |
+
downloadLink.style.display = 'none';
|
12 |
+
audioPlayer.style.display = 'none';
|
13 |
+
|
14 |
+
try {
|
15 |
+
const response = await fetch('http://localhost:5000/text-to-speech/', {
|
16 |
+
method: 'POST',
|
17 |
+
headers: {
|
18 |
+
'Content-Type': 'application/json',
|
19 |
+
},
|
20 |
+
body: JSON.stringify({ text: text }),
|
21 |
+
});
|
22 |
+
|
23 |
+
const data = await response.json();
|
24 |
+
|
25 |
+
if (data.status === 'success') {
|
26 |
+
statusDiv.textContent = 'Speech generated successfully!';
|
27 |
+
downloadLink.href = 'http://localhost:5000' + data.url;
|
28 |
+
downloadLink.style.display = 'block';
|
29 |
+
audioPlayer.src = 'http://localhost:5000' + data.url;
|
30 |
+
audioPlayer.style.display = 'block';
|
31 |
+
} else {
|
32 |
+
statusDiv.textContent = `Error: ${data.message}`;
|
33 |
+
}
|
34 |
+
} catch (error) {
|
35 |
+
statusDiv.textContent = `Network error: ${error}`;
|
36 |
+
}
|
37 |
+
});
|
38 |
+
});
|
speaker_reference.wav
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e7fa2e3b58516ba6057ab93fa819eca83097c31a90d24e53cc6593ef384ce1f1
|
3 |
+
size 188476
|
start.sh
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
#!/bin/bash
|
2 |
+
|
3 |
+
# 1. Navigate to the application directory (where local_server_new.py is)
|
4 |
+
cd /app
|
5 |
+
|
6 |
+
# 2. Activate the Python virtual environment
|
7 |
+
source venv/bin/activate
|
8 |
+
|
9 |
+
# 3. Run the FastAPI application using Uvicorn
|
10 |
+
# - Bind to all interfaces (0.0.0.0)
|
11 |
+
# - Listen on port 80 (required by Hugging Face Spaces for HTTP)
|
12 |
+
# - Specify your FastAPI application module and app instance (local_server_new:app)
|
13 |
+
uvicorn local_server_new:app --host 0.0.0.0 --port 80
|
style.css
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
body {
|
2 |
+
font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
|
3 |
+
background-color: #f4f7f6;
|
4 |
+
margin: 40px;
|
5 |
+
display: flex;
|
6 |
+
flex-direction: column;
|
7 |
+
align-items: center;
|
8 |
+
color: #333;
|
9 |
+
}
|
10 |
+
|
11 |
+
h1 {
|
12 |
+
color: #2c3e50;
|
13 |
+
margin-bottom: 30px;
|
14 |
+
text-align: center;
|
15 |
+
font-size: 2.5em;
|
16 |
+
}
|
17 |
+
|
18 |
+
textarea {
|
19 |
+
padding: 15px;
|
20 |
+
border: 1px solid #ccc;
|
21 |
+
border-radius: 8px;
|
22 |
+
font-size: 1em;
|
23 |
+
margin-bottom: 20px;
|
24 |
+
width: 80%;
|
25 |
+
max-width: 600px;
|
26 |
+
box-sizing: border-box;
|
27 |
+
resize: vertical; /* Allows vertical resizing */
|
28 |
+
box-shadow: 2px 2px 10px rgba(0, 0, 0, 0.1);
|
29 |
+
}
|
30 |
+
|
31 |
+
button {
|
32 |
+
background-color: #3498db;
|
33 |
+
color: white;
|
34 |
+
padding: 12px 25px;
|
35 |
+
border: none;
|
36 |
+
border-radius: 8px;
|
37 |
+
cursor: pointer;
|
38 |
+
font-size: 1.1em;
|
39 |
+
transition: background-color 0.3s ease;
|
40 |
+
box-shadow: 2px 2px 5px rgba(0, 0, 0, 0.15);
|
41 |
+
}
|
42 |
+
|
43 |
+
button:hover {
|
44 |
+
background-color: #2980b9;
|
45 |
+
}
|
46 |
+
|
47 |
+
#status {
|
48 |
+
margin-top: 20px;
|
49 |
+
font-weight: bold;
|
50 |
+
color: #27ae60; /* Green for success, you can change for errors */
|
51 |
+
}
|
52 |
+
|
53 |
+
#audioOutput {
|
54 |
+
margin-top: 30px;
|
55 |
+
text-align: center;
|
56 |
+
}
|
57 |
+
|
58 |
+
#downloadLink {
|
59 |
+
display: inline-block;
|
60 |
+
background-color: #2ecc71;
|
61 |
+
color: white;
|
62 |
+
padding: 10px 20px;
|
63 |
+
border-radius: 5px;
|
64 |
+
text-decoration: none;
|
65 |
+
font-size: 1em;
|
66 |
+
transition: background-color 0.3s ease;
|
67 |
+
box-shadow: 2px 2px 5px rgba(0, 0, 0, 0.15);
|
68 |
+
}
|
69 |
+
|
70 |
+
#downloadLink:hover {
|
71 |
+
background-color: #27ae60;
|
72 |
+
}
|
73 |
+
|
74 |
+
#audioPlayer {
|
75 |
+
margin-top: 10px;
|
76 |
+
}
|
web/index.html
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<!DOCTYPE html>
|
2 |
+
<html>
|
3 |
+
<head>
|
4 |
+
<title>Text to Speech</title>
|
5 |
+
<link rel="stylesheet" href="style.css">
|
6 |
+
</head>
|
7 |
+
<body>
|
8 |
+
<h1>Text to Speech</h1>
|
9 |
+
<textarea id="inputText" rows="5" cols="50"></textarea><br><br>
|
10 |
+
<button id="convertButton">Convert to Speech</button>
|
11 |
+
<div id="status"></div>
|
12 |
+
<div id="audioOutput" style="margin-top: 20px;">
|
13 |
+
<a id="downloadLink" href="#" download="output.wav" style="display: none;">Download Audio</a>
|
14 |
+
<audio id="audioPlayer" controls style="display: none;"></audio>
|
15 |
+
</div>
|
16 |
+
|
17 |
+
<script src="script.js"></script>
|
18 |
+
</body>
|
19 |
+
</html>
|
web/script.js
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
document.addEventListener('DOMContentLoaded', () => {
|
2 |
+
const convertButton = document.getElementById('convertButton');
|
3 |
+
const inputText = document.getElementById('inputText');
|
4 |
+
const statusDiv = document.getElementById('status');
|
5 |
+
const downloadLink = document.getElementById('downloadLink');
|
6 |
+
const audioPlayer = document.getElementById('audioPlayer');
|
7 |
+
|
8 |
+
convertButton.addEventListener('click', async () => {
|
9 |
+
const text = inputText.value;
|
10 |
+
statusDiv.textContent = 'Processing...';
|
11 |
+
downloadLink.style.display = 'none';
|
12 |
+
audioPlayer.style.display = 'none';
|
13 |
+
|
14 |
+
try {
|
15 |
+
const response = await fetch('http://localhost:5000/text-to-speech/', {
|
16 |
+
method: 'POST',
|
17 |
+
headers: {
|
18 |
+
'Content-Type': 'application/json',
|
19 |
+
},
|
20 |
+
body: JSON.stringify({ text: text }),
|
21 |
+
});
|
22 |
+
|
23 |
+
const data = await response.json();
|
24 |
+
|
25 |
+
if (data.status === 'success') {
|
26 |
+
statusDiv.textContent = 'Speech generated successfully!';
|
27 |
+
downloadLink.href = 'http://localhost:5000' + data.url;
|
28 |
+
downloadLink.style.display = 'block';
|
29 |
+
audioPlayer.src = 'http://localhost:5000' + data.url;
|
30 |
+
audioPlayer.style.display = 'block';
|
31 |
+
} else {
|
32 |
+
statusDiv.textContent = `Error: ${data.message}`;
|
33 |
+
}
|
34 |
+
} catch (error) {
|
35 |
+
statusDiv.textContent = `Network error: ${error}`;
|
36 |
+
}
|
37 |
+
});
|
38 |
+
});
|
web/style.css
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
body {
|
2 |
+
font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
|
3 |
+
background-color: #f4f7f6;
|
4 |
+
margin: 40px;
|
5 |
+
display: flex;
|
6 |
+
flex-direction: column;
|
7 |
+
align-items: center;
|
8 |
+
color: #333;
|
9 |
+
}
|
10 |
+
|
11 |
+
h1 {
|
12 |
+
color: #2c3e50;
|
13 |
+
margin-bottom: 30px;
|
14 |
+
text-align: center;
|
15 |
+
font-size: 2.5em;
|
16 |
+
}
|
17 |
+
|
18 |
+
textarea {
|
19 |
+
padding: 15px;
|
20 |
+
border: 1px solid #ccc;
|
21 |
+
border-radius: 8px;
|
22 |
+
font-size: 1em;
|
23 |
+
margin-bottom: 20px;
|
24 |
+
width: 80%;
|
25 |
+
max-width: 600px;
|
26 |
+
box-sizing: border-box;
|
27 |
+
resize: vertical; /* Allows vertical resizing */
|
28 |
+
box-shadow: 2px 2px 10px rgba(0, 0, 0, 0.1);
|
29 |
+
}
|
30 |
+
|
31 |
+
button {
|
32 |
+
background-color: #3498db;
|
33 |
+
color: white;
|
34 |
+
padding: 12px 25px;
|
35 |
+
border: none;
|
36 |
+
border-radius: 8px;
|
37 |
+
cursor: pointer;
|
38 |
+
font-size: 1.1em;
|
39 |
+
transition: background-color 0.3s ease;
|
40 |
+
box-shadow: 2px 2px 5px rgba(0, 0, 0, 0.15);
|
41 |
+
}
|
42 |
+
|
43 |
+
button:hover {
|
44 |
+
background-color: #2980b9;
|
45 |
+
}
|
46 |
+
|
47 |
+
#status {
|
48 |
+
margin-top: 20px;
|
49 |
+
font-weight: bold;
|
50 |
+
color: #27ae60; /* Green for success, you can change for errors */
|
51 |
+
}
|
52 |
+
|
53 |
+
#audioOutput {
|
54 |
+
margin-top: 30px;
|
55 |
+
text-align: center;
|
56 |
+
}
|
57 |
+
|
58 |
+
#downloadLink {
|
59 |
+
display: inline-block;
|
60 |
+
background-color: #2ecc71;
|
61 |
+
color: white;
|
62 |
+
padding: 10px 20px;
|
63 |
+
border-radius: 5px;
|
64 |
+
text-decoration: none;
|
65 |
+
font-size: 1em;
|
66 |
+
transition: background-color 0.3s ease;
|
67 |
+
box-shadow: 2px 2px 5px rgba(0, 0, 0, 0.15);
|
68 |
+
}
|
69 |
+
|
70 |
+
#downloadLink:hover {
|
71 |
+
background-color: #27ae60;
|
72 |
+
}
|
73 |
+
|
74 |
+
#audioPlayer {
|
75 |
+
margin-top: 10px;
|
76 |
+
}
|